全基因组测序分析流程
2020-10-18 本文已影响0人
一只烟酒僧
全代码
#!/bin/bash
#用户目录及预定义变量
wk_dir=/media/whq/282A932A2A92F3D2/282A932A2A92F3D2/WHQ/xxx_xxx_NGS
code_dir=/media/whq/282A932A2A92F3D2/WHQ/myRscript
fastq_dir=/media/whq/282A932A2A92F3D2/282A932A2A92F3D2/WHQ/xxx_xxx_NGS/fastq
index=$wk_dir/index/gatk_xxxx.fasta
genome=$wk_dir/genomexxxxx38.fasta
known_site_dir=$wk_dir/index/known_sites
dir_prefix=FDSW*
fastq1_suffix=_1.fastq.gz
fastq2_suffix=_2.fastq.gz
#BWA+sort
cd $wk_dir/fastq
for id in `ls -d $dir_prefix`
do
{
input1=$id/$id$fastq1_suffix;input2=$id/$id$fastq2_suffix;sample=$id;index=$index;bwa mem -t 15 -R "@RG\tID:$sample\tSM:$sample\tLB:WGS\tPL:Illumina" $index $input1 $input2|samtools sort -@ 10 -o $id/$id.sorted.bam -
} 2>$wk_dir/temp/$id'.bwa.sorted.errlog' &
done
wait
#mark_dup+BQSR
cd $wk_dir/fastq
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'.sorted.bam';output=$id/$id'.sorted.markdup.bam';gatk MarkDuplicates -I $input -O $output -M $id/$id'.markdup.matrix'
} 2>$wk_dir/temp/$id'.markdup.errlog' &
done
wait
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'.sorted.markdup.bam' ;output=$id/$id'.bqsr.table';ref=$genome;know_site1=$known_site_dir/1000G_phase1.snps.high_confidence.hg38.vcf.gz;know_site2=$known_site_dir/Mills_and_1000G_gold_standard.indels.hg38.vcf.gz;know_site3=$known_site_dir/dbsnp_138.hg38.vcf.gz;gatk BaseRecalibrator -I $input -O $output -R $ref --known-sites $know_site1 --known-sites$know_site2 --known-sites $know_site3
} 2>$wk_dir/temp/$id'.BQSR.1.errlog' &
done
wait
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'.sorted.markdup.bam';output=$id/$id'.sorted.markdup.BQSR.bam';ref=$genome;bqsr=$id/$id'.bqsr.table';gatk ApplyBQSR -R $ref -I $input -O $output -bqsr $bqsr
} 2>$wk_dir/temp/$id'.BQSR.2.errlog' &
done
wait
#call SNP+INDEL
cd $wk_dir/fastq
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'.sorted.markdup.BQSR.bam';output=$id/$id'_raw.vcf';dbsnp=$known_site_dir/dbsnp_138.hg38.vcf.gz;ref=$genome;gatk HaplotypeCaller -R $ref -I $input -O $output --dbsnp $dbsnp
} 2>$wk_dir/temp/$id'.call.snpindel.errlog' &
done
wait
#gatk分离SNP和indel
cd $wk_dir/fastq
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'_raw.vcf'
output=$id/$id'_snp.vcf'
gatk SelectVariants -select-type SNP -V $input -O $output
} 2>$wk_dir/temp/$id'.selectvariants.snp.errlog' &
done
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'_raw.vcf'
output=$id/$id'_indel.vcf';
gatk SelectVariants -select-type INDEL -V $input -O $output
} 2>$wk_dir/temp/$id'.selectvariants.indel.errlog' &
done
wait
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'_snp.vcf'
output=$id/$id'_snp.filtered.vcf'
gatk VariantFiltration -V $input --filter-expression "DP<4 || QD < 2.0 || MQ < 40.0 || FS > 60.0 || SOR > 3.0 || MQRankSum < -12.5 || ReadPosRankSum < -8.0" --filter-name "Filter" -O $output
} 2>$wk_dir/temp/$id'.hardfilter.snp.errlog' &
done
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'_indel.vcf'
output=$id/$id'_indel.filtered.vcf'
gatk VariantFiltration -V $input --filter-expression "DP<4 || QD < 2.0 || FS > 200.0 || SOR > 10.0 || MQRankSum < -12.5 || ReadPosRankSum < -8.0" --filter-name "Filter" -O $output
} 2>$wk_dir/temp/$id'.hardfilter.indel.errlog' &
done
wait
for id in `ls -d $dir_prefix`
do
{
input1=$id/$id'_snp.filtered.vcf'
input2=$id/$id'_indel.filtered.vcf'
output=$id/$id'.filtered.vcf'
gatk MergeVcfs -I $input1 -I $input2 -O $output
} 2>$wk_dir/temp/$id'.mergevcf.errlog' &
done
各部分细节
一、准备工作
#需要的文件
#1、SRA库中下载的SRRid（文件）#保存到temp中,命名为srrlist
#2、从ENA库中爬取链接的脚本 #原文件在/media/whq/282A932A2A92F3D2/WHQ/myRscript，保存到code目录下
#!/bin/bash
#创建文件系统
cd $wk_dir
mkdir fastq #存储原始数据及各个阶段的数据
mkdir code#用于存储代码
mkdir index #存储索引数据
mkdir genome #存储基因组fasta数据
mkdir res #存储最终结果数据
mkdir temp #存储中间的temp数据
mkdir otherfile #存储其它过程中间数据
#将构建好的index和基因组分别放到index和genome中
#用户目录及预定义变量
wk_dir=/media/whq/282A932A2A92F3D2/282A932A2A92F3D2/WHQ/xxx_xxxx_NGS
code_dir=/media/whq/282A932A2A92F3D2/WHQ/myRscript
fastq_dir=/media/whq/282A932A2A92F3D2/282A932A2A92F3D2/WHQ/xxx_xxxx_NGS/fastq
index=$wk_dir/index/gatk_xxxx_xxxx.fasta
genome=$wk_dir/genome/xxxx.fasta
known_site_dir=$wk_dir/index/known_sites
dir_prefix=FDSW*
fastq1_suffix=_1.fastq.gz
fastq2_suffix=_2.fastq.gz
二、下载文件(如果是自己测得数据，可以直接将数据放到fastq文件夹中即可)
#下载文件
cp $code_dir/get_fastq_data_from_ena.sh $code_dir/get_fastq_url_from_ena.R $wk_dir/code #将爬取连接和下载的脚本拷贝到code下
nohup $wk_dir/code/get_fastq_data_from_ena.sh $wk_dir/temp/srrlist $fastq_dir 1> download_ena_data.log 2>download_ena_data.errlog & #传的第一个参数是srrlist，第二个参数是fastq保存的dir
三、整理文件
#整理文件
cd $wk_dir/fastq
ls SRR*_1*|while read id;do dir=${id%%_*.fastq.gz};mkdir $dir;mv $dir* $dir;done
四、比对、sort
#参考基因组和索引
cd $wk_dir/index
ln -s ../../reference_genome/ #建立基因组文件的软链接
ln -s ../../reference_index/ #建立索引文件的软链接
#比对、排序
cd $wk_dir/fastq
ls -d SRR*|while read id;do input1=$id/$id'_1.fastq.gz';input2=$id/$id'_2.fastq.gz';index=$wk_dir/index/genome_index/gatk_human_hg38.fasta;bwa mem -t 15 -R "@RG\tID:$id\tSM:$id\tLB:WGS\tPL:Illumina" $index $input1 $input2|samtools sort -@ 15 -o $id/$id.sorted.bam - ;done 1>$wk_dir/temp/mapping.log 2>$wk_dir/temp/mapping.errlog
五、标记重复和BQSR
#mark_dup+BQSR
cd $wk_dir/fastq
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'.sorted.bam';output=$id/$id'.sorted.markdup.bam';gatk MarkDuplicates -I $input -O $output -M $id/$id'.markdup.matrix'
} 2>$wk_dir/temp/$id'.markdup.errlog' &
done
wait
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'.sorted.markdup.bam' ;output=$id/$id'.bqsr.table';ref=$genome;know_site1=$known_site_dir/1000G_phase1.snps.high_confidence.hg38.vcf.gz;know_site2=$known_site_dir/Mills_and_1000G_gold_standard.indels.hg38.vcf.gz;know_site3=$known_site_dir/dbsnp_138.hg38.vcf.gz;gatk BaseRecalibrator -I $input -O $output -R $ref --known-sites $know_site1 --known-sites$know_site2 --known-sites $know_site3
} 2>$wk_dir/temp/$id'.BQSR.1.errlog' &
done
wait
六、call SNP和INDEL
#call SNP+INDEL
cd $wk_dir/fastq
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'.sorted.markdup.BQSR.bam';output=$id/$id'_raw.vcf';dbsnp=$known_site_dir/dbsnp_138.hg38.vcf.gz;ref=$genome;gatk HaplotypeCaller -R $ref -I $input -O $output --dbsnp $dbsnp
} 2>$wk_dir/temp/$id'.call.snpindel.errlog' &
done
wait
七、gatk提取SNP和INDEL
#gatk分离SNP和indel
cd $wk_dir/fastq
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'_raw.vcf'
output=$id/$id'_snp.vcf'
gatk SelectVariants -select-type SNP -V $input -O $output
} 2>$wk_dir/temp/$id'.selectvariants.snp.errlog' &
done
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'_raw.vcf'
output=$id/$id'_indel.vcf';
gatk SelectVariants -select-type INDEL -V $input -O $output
} 2>$wk_dir/temp/$id'.selectvariants.indel.errlog' &
done
八、使用gatk推荐的参数进行硬过滤
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'_snp.vcf'
output=$id/$id'_snp.filtered.vcf'
gatk VariantFiltration -V $input --filter-expression "DP<4 || QD < 2.0 || MQ < 40.0 || FS > 60.0 || SOR > 3.0 || MQRankSum < -12.5 || ReadPosRankSum < -8.0" --filter-name "Filter" -O $output
} 2>$wk_dir/temp/$id'.hardfilter.snp.errlog' &
done
for id in `ls -d $dir_prefix`
do
{
input=$id/$id'_indel.vcf'
output=$id/$id'_indel.filtered.vcf'
gatk VariantFiltration -V $input --filter-expression "DP<4 || QD < 2.0 || FS > 200.0 || SOR > 10.0 || MQRankSum < -12.5 || ReadPosRankSum < -8.0" --filter-name "Filter" -O $output
} 2>$wk_dir/temp/$id'.hardfilter.indel.errlog' &
done
wait
for id in `ls -d $dir_prefix`
do
{
input1=$id/$id'_snp.filtered.vcf'
input2=$id/$id'_indel.filtered.vcf'
output=$id/$id'.filtered.vcf'
gatk MergeVcfs -I $input1 -I $input2 -O $output
} 2>$wk_dir/temp/$id'.mergevcf.errlog' &
done
